English (pdf)
Article in xml format
Article references
Automatic translation
Send this article by e-mail
Cited by SciELO 
Cited by Google 
Similars in
SciELO 
Similars in Google 
Permalink
Phosphorus (P) deficiency is one of the most widespread and economically important mineral deficiencies affecting grazing livestock. For example, based on soil P concentrations, some 70% of northern Australian rangeland pastures are expected to be deficient in P for grazing cattle (McCosker and Winks 1994; CSIRO 2007). P deficiency reduces voluntary pasture intake, leading to adverse effects on liveweight (LW) change and hence live weight in both growing and breeder cattle, lower calving rates and calf growth, and increased breeder mortality (Winks 1990; McCosker and Winks 1994; Suttle 2010). The economic benefits of P supplementation (Dixon et al. 2011) obviously depend on the magnitude and consistency of animal responses in specific regions and paddocks and across years.
Large between-year variability in the amount and distribution of rainfall is typical of many regions in the seasonally dry tropics, and particularly in northern Australia (Weston 1988; Cook and Heerdegen 2001). As a consequence there is generally high between-year variability in pasture quantity and quality (McLennan et al. 1988) and in annual LW gains of growing cattle (McCown 1981; Winks 1984; Jones et al. 1990). The performance of breeder cows in terms of LW, fertility and mortality also varies between years in association with climatic factors. However, the effects of environmental variability on responses to P supplementation in LW and productivity of breeder herds grazing pastures growing on P-deficient soils, and the economic implications of these variable responses, have received little attention. Furthermore, between-year variability in responses by breeder cows to P supplementation is often complicated by carry-over effects through successive annual cycles associated with cow body condition, time of calving and skeletal P reserves (Dixon et al. 2017). The present study examined the magnitude of the variation between years in performance and the responses to providing adequate dietary P in groups of breeder cows grazing pastures growing on severely P-deficient soil at a site in the seasonally dry tropics.
Materials and Methods
General outline and animal cohorts
The study reports results from experiments with breeder cows at the Lansdown Pasture Research Station (19º41' S, 146º51' E), approximately 50 km south of Townsville in north Queensland, Australia. Droughtmaster breeders (Bos indicus × Bos taurus) grazed an area of pasture growing on soils very low in available P through 4 annual cycles in Experiment 1 (Expt 1), and during an additional fifth annual cycle, when cows grazed similar low-P pasture in an adjacent area (Expt 2). The soils of the experimental site comprised a mixture of yellow earth (Gn 2.64, Northcote 1971), solidic (Dy 3.43) and solodic/solonized-solonetz (Dy 3.43) types (Murtha and Crack 1966; Coates 1994). The low dietary-P treatment (Pdefic) was imposed by grazing the cattle on the P-deficient pasture without any additional dietary P. The P-supplemented treatment (Psupp) was achieved by grazing the cattle on the same P-deficient pasture and providing a P supplement during Expt 1 or by grazing pasture fertilized with P in Expt 2. Since each new group of heifers or cows used in the study (i.e. Expts 1a, 1b and 2) had grazed pastures expected to be adequate in P for at least 2 years before introduction to the experimental treatments, it was expected that the animals would initially have had replete body P reserves. Rainfall was recorded at the Lansdown weather station (Bureau of Meteorology Station #33226) c. 2 km from the experimental site.
Experiment 1
The trial paddocks were in an area of P-deficient soil with bicarbonate extractable P (PB, Colwell 1963) of c. 4 ppm in the top 100 mm of soil and comprised part of the P-deficient treatment paddock described by Coates et al. (2018). The pasture was a mixture of native grasses (Heteropogon contortus, Chrysopogon fallax, Aristida spp. and naturalized Bothriochloa pertusa) plus sown sabi grass (Urochloa mosambicensis cv. Nixon) and sown legumes (Stylosanthes hamata cv. Verano and S. scabra cv. Seca). The 16 ha trial area was divided into 2 paddocks of 8 ha. Each was grazed by a group of 4 breeders, which were allocated to the treatments by stratified randomization according to LW. One group (Psupp group) received a P supplement via medication of the drinking water with Monophos (NaH2PO4) to provide 0.2 g P/L water. The Pdefic treatment received no P supplement and both groups had access to salt blocks (sodium chloride). The treatment groups were alternated between the 2 paddocks fort-nightly to minimize between-paddock effects. Access to water was arranged to maintain the integrity of the treatments.
Measurements were made throughout 4 annual grazing cycles (Expts 1a, 1b, 1c and 1d). Expt 1a commenced in September 1986 with 3.5-year-old cows pregnant with their second calf (second-calf cows, SCC; calving October‒December) and continued until June 1987 when the paddocks were destocked for animal welfare reasons owing to severe drought. The paddocks remained de-stocked until August 1988, when a new draft of 8 pregnant heifers, initially c. 2.5-year-old, (first-calf cows, FCC; calving October‒December) was allocated to the 2 treatments (Psupp and Pdefic) by stratified randomization according to LW. These groups of animals remained in the study and received the same treatments through 3 annual cycles: 1988-89 (Expt 1b, FCC), 1989-90 (Expt 1c, SCC; calving December-January) and 1990-91 (Expt 1d, mature cows; calving December-January). The only exception was that one non-pregnant cow in the Psupp group was replaced by a P-replete pregnant cow at the end of the Expt 1c grazing cycle. In Expts 1a, 1b and 1c the cows were joined for at least 3 months with 1 physically sound bull per treatment group after calving was complete. Calves were weaned in early to mid-June of each grazing cycle.
Experiment 2
In Expt 2 measurements were made on Pdefic and Psupp groups of FCC which initially were pregnant heifers c. 2.5 years old in April 1994. The Pdefic heifers (n = 10) grazed a 20 ha paddock that comprised the area used in Expt 1 plus an adjoining 4 ha of similar low-P pasture. The Psupp heifers (n = 10) received no P supplement directly but grazed a nearby 16 ha sabi grass-stylo pasture which had been fertilized annually for >10 years with super- phosphate at 10 kg P/ha so that improved dietary P was provided by the pasture. This experiment has been described previously (Coates et al. 2018), where the focus was primarily on the mineral density of tail bone. The heifers calved in November-December 1994 and the calves were weaned on 11 May 1995, when the study was terminated. Both groups of breeders were joined with one bull per group for 3 months after calving was completed.
Measurements
Cows and calves were weighed regularly following an overnight fast, usually at about 4 week intervals. Jugular blood samples from the cows were collected on most weigh days and centrifuged to separate plasma, which was retained for subsequent measurement of plasma inorganic phosphorus (PIP) concentrations. PIP was measured at 4‒8 week intervals in Expt 1a, predominantly at 4 week intervals in Expts 1b, 1c and 1d, and at about 8 week intervals in Expt 2, and analyses were as described by Murphy and Riley (1962). Samples of feces were obtained from each cow per rectum when the cows were weighed during Expts 1b, 1c and 1d. The δ13C in the individual samples of feces was measured by mass spectrometry (LeFeuvre and Jones 1988) and used to calculate the proportion of C3 plants (i.e. non-grass components, comprising predominantly Stylosanthes spp.) and C4 tropical grasses in the diet (Jones et al. 1979). In addition the concentrations of total P and total N were measured in feces sampled during Expt 1b and part of Expt 1c. Pregnancy rates were determined at weaning by manual palpation of the uterus via the rectum by an experienced operator to select pregnant animals for each new group, and also of the cows at weaning at the end of each annual cycle.
Animal welfare
All experimental procedures involving the animals were carried out according to the Code of Practice for the Care and Use of Animals for Scientific Purposes and with the approval of the relevant Animal Ethics Committees operating when the experiments were conducted.
Calculations and statistical analyses
Statistical analysis within each grazing cycle was restricted to cows that reared a calf to weaning because of the effects of pregnancy and lactation on cow LW and PIP. Statistical analysis was conducted using GENSTAT (release 16.1 9VSN International Ltd, Hemel Hemstead, UK). There was no paddock replication in either experiment and the differences in measured parameters between the Pdefic and Psupp groups of cows at each measurement date were examined using ANOVA with individual animals considered as the experimental units. As the Pdefic and Psupp treatment groups were rotated between paddocks each fortnight in Expt 1, we considered the effects of any paddock differences on response parameters were minimized. The LW of the cows at the beginning of each grazing cycle, and the individual calving dates, were examined as covariates in the analyses of response variables, and included when there was a greater than 90% probability that the covariates did affect the responses.
Results
Seasonal conditions
Monthly rainfall for the grazing years 1986-87 to 1990-91 (Expt 1) and for 1994-95 (Expt 2) and the dates of the seasonal breaks (the first 3 day interval after June 30 when ≥50 mm rain was received) are shown in Table 1. These data highlight the summer-dominant rainfall pattern and the extreme variability in both total and effective annual rainfall in this environment. The 1986-87 grazing year (Expt 1a) ended in severe drought with insufficient rainfall during any 3 day interval to constitute a seasonal break as defined herein. Rainfall during the 6 months January‒June 1987 was only 45% of the long-term mean for that interval and the trial area was destocked during the 1987-88 grazing cycle. In contrast, rainfall for 1988-89 and 1989-90 was high at 128 and 129% of the long-term annual average, while that for 1990-91 was excessive (197% of the long-term average) with particularly high rainfall during January-February 1991. In 1994-95 (Expt 2) rainfall was only 36% of the long-term annual average.
Cow LW and LW change
Mean LWs of cows through each grazing cycle are presented in Figure 1. Differences between grazing cycles in the effects of P treatment on cow LWs and patterns of LW change were unexpectedly large due most likely to the large differences between grazing cycles in the amounts and distribution of rainfall. In Expt 1a, when rainfall was much lower than average (Table 1), the differences in LW between Pdefic and Psupp cows increased progressively from the end of November (P<0.05). By the end of the annual cycle the Psupp cows had gained 30 kg whereas Pdefic cows had lost 62 kg, i.e. a difference of 92 kg, which was significant (P<0.01). Treatment effects on cow LWs during Expt 1b (1988-89) were in marked contrast to those in Expt 1a. Rainfall was well above average from March to June resulting in an extended green season. In addition, on average, the Psupp cows calved 28 days earlier than the Pdefic cows so were lactating for longer during the grazing cycle; this is the likely reason for the lower mean LW of the Psupp cows than that of the Pdefic cows, although these differences failed to reach significance (P>0.05). In Expt 1c (1989-90) rainfall was comparable with that in Expt 1a until the end of February 1990, but there was abnormally high rainfall in March and April (346 and 337 mm, respectively). The differences in mean LW between Pdefic and Psupp cows became progressively greater until mid-January 1990, when Psupp cows were 59 kg heavier than Pdefic cows. From mid-January until the end of the grazing cycle in June 1990, LW changes in both groups were similar and the final LW advantage of Psupp cows was 49 kg. When initial LW was used as a covariate in the analysis, treatment differences were significant on all weighing dates from August 1989 to March 1990 (P<0.05 to <0.001). When calf birth date was used as a covariate, treatment differences in cow LWs were significant from January to June 1990 (P<0.05 to <0.01). During the full annual cycle Psupp cows gained 43 kg and Pdefic cows lost 5 kg with the difference approaching significance (P=0.07). There was a carryover effect on cow LW (Figure 1) from Expt 1c to Expt 1d such that the average LW of Psupp cows was initially 33 kg greater than that of Pdefic cows. Changes in LWs during the Expt 1d grazing cycle were similar for the 2 treatment groups. When initial LW was included as a covariate in the analysis, differences were not significant (P=0.14 to 0.89) except at the end of the grazing cycle in June and July 1991 (P=0.03). Through the entire grazing cycle, Psupp cows gained 17 kg, while Pdefic cows lost 5 kg and the difference (22 kg) was significant (P=0.01).
Table 1 Rainfall (mm) during each of the annual cycles for Expts 1 and 2. In Experiment 1 the area was destocked from 10 June 1987 until 4 August 1988 (1987‒88). The long-term (1891‒1990) average rainfall (BoM 2019) is also given. The seasonal break was calculated as the first 3 day interval after June 30 when ≥50 mm rain was received.
1There was not sufficient rain during any 3 day interval to meet the designated criterion for a seasonal break.
Figure 1a Treatment mean live weights (LW) of cows grazing P-deficient pasture without (○) or with (●) P supplement through 4 annual cycles (1986-87, 1988-89, 1989-90 and 1990-91), or grazing P-deficient or P-fertilized pasture in 1994-95. Figure 1b. Plasma inorganic phosphorus (PIP) concentrations of cows without (Δ) or with (▲) P supplementation or P-fertilizer application through the same annual cycles. Cows in Expt 1 calved in November-December and were weaned in June, while those in Expt 2 calved in December-January and were weaned in mid-May.
In Expt 2 there were no significant differences in LWs between Pdefic and Psupp treatment groups between April and December 1994 by which time calving was complete. Thereafter Psupp cows gained LW rapidly while Pdefic cows virtually maintained LW. The differences between mean LWs of the 2 groups became increasingly greater (P<0.01 in February 1995 and P<0.001 thereafter). At the end of the annual cycle Pdefic cows had lost 71 kg LW compared with a 44 kg gain in the Psupp group (i.e. a difference of 115 kg LW).
Plasma inorganic phosphorus (PIP) profiles
There were large differences between annual cycles in the effects of treatment on PIP profiles (Figure 1b), with differences apparently related to seasonal rainfall patterns in a similar manner to the differences in the LW profiles. In Expt 1a there were large treatment differences after calving (P<0.01) when all cows were lactating. Mean PIP of the Pdefic cows was only 22 mg/L from December to June, reflecting the effects of low soil P, probably exacerbated by the effects of low rainfall on forage P concentration and the high demand for P during lactation. Over the same interval PIP of Psupp cows averaged 57 mg/L, reflecting the increased P intake in this group through supplementation and likely also increased pasture intake. In Expt 1b calving was complete in both groups by the end of December. The PIP of Pdefic cows for the 6 sampling occasions from December to May averaged 42 mg/L compared with 54 mg/L for the Psupp cows, but on no sampling occasion was the difference between groups significant (P>0.10). Mean PIP concentrations for all sampling dates in Expt 1c were consistently higher in the Psupp treatment than in the Pdefic group, the differences being significant on all 12 occasions (P<0.05 to P<0.001). During lactation (December‒June), PIP of Pdefic cows averaged only 26 mg/L, half that of Psupp cows (50 mg/L). In Expt 1d, fewer samplings were carried out and PIP in Pdefic cows remained appreciably lower than in Psupp cows for most of the grazing cycle; differences between mean values for the 2 groups between August 1990 and April 1991 were significant on 3 occasions (P=0.002 to 0.01) and approached significance on 2 occasions (P=0.08). For the 3 sampling occasions conducted during late lactation (March‒May 1991) mean PIP of cows in the Pdefic treatment (20 mg/L) was less than half that of cows in the Psupp treatment (44 mg/L). In Expt 2 the PIP in Pdefic cows for the period July 1994‒June 1995 averaged only one-third of the PIP of Psupp cows (14 vs. 53 mg/L; P<0.01; Figure 1b) and indicated severe and prolonged deficiency of dietary P.
Contribution of Stylosanthes to the diet and fecal concentrations of P and N
During the early to mid-wet season (November-January) in Expts 1b, 1c and 1d, when measurements were made, stylo comprised c. 10‒30% of the diet selected by both treatment groups of cows (Figure 2). However the proportion of stylo increased progressively from the mid-wet season through to the mid-dry season (February to July), and remained as a high proportion through to the late dry season (August-September). Furthermore, during these latter intervals the Psupp cows increased their selection of stylo to a greater extent (to c. 70‒85%) than the Pdefic cows (c. 40‒60%).
Figure 2 The percentage of non-grass (i.e. principally Stylosanthes spp.) in the diet selected by cows not fed P supplement (□) or provided with P supplement (■) from 30 August 1988 until 2 July 1991 during Expts 1b, 1c and 1d. The non-grass in the diet was calculated from the measured 13δC of feces.
During the interval from 30 August 1988 to 13 February 1990 (Expt 1b and part of Expt 1c), when measurements were made, the concentration of P in feces of Pdefic cows averaged 1.93 g P/kg DM, whereas the concentration in Psupp cows was consistently higher and averaged 2.41 g P/kg DM (Figure 3a). During Expt 1b fecal P concentration was higher during the wet season (December-April; >2.2 g P/kg DM) than during the following dry season (May‒November; 1.5-2.1 g P/kg DM). During the late dry season fecal N concentration was generally 12‒14 g N/kg DM in both treatment groups (Figure 3b). However fecal N was generally 14‒20 g N/kg DM during the wet season through to the mid dry season (December-July) of Expt 1b, and on average tended to be higher for Psupp cows (mean 17.2 g N/kg DM) than for Pdefic cows (mean 15.1 g N/kg DM).
Figure 3a The concentration of P in feces of cows not fed a P supplement (Δ) or provided with P supplement (▲) from 30 August 1988 until 13 February 1990 during Expts 1b and 1c. Figure 3b. The concentration of N in feces of cows not fed a P supplement (○) or provided with a P supplement (●) during the same interval.
Calf growth rates and reconception in the cows
Calf growth rates (Table 2) averaged 0.77 kg/d in the Pdefic treatment groups across the 5 grazing cycles. Calf growth rate was on average 0.09 kg/day higher in the Psupp treatment groups, but the difference was significant (P<0.05) only for Expt 1a. In Expt 1a, none of the Pdefic cows, but 3 of the 4 Psupp cows, conceived. In Expt 1b, all Pdefic cows and 3 of the 4 Psupp cows conceived. The cow that failed to conceive in the Psupp group was comparable in LW and condition with the cows that conceived. All Pdefic and Psupp cows conceived during Expt 1c. Cows were not joined in Expt 1d. In Expt 2, 9 of the 10 Psupp cows conceived, while only 1 of the 10 Pdefic cows conceived.
Discussion
The most important findings from the study were the large variations between annual cycles in: (i) the performance of breeder cows grazing P-deficient pastures; and (ii) the responses of the breeder cows to the provision of additional dietary P via supplement or fertilizer. The differences between annual cycles in the responses to improved P intake were observed in cow LWs and LW change profiles, PIP profiles, cow conception rates and to some extent calf growth rate, and were associated with large variation in the amount and pattern of rainfall during each annual cycle as shown in Table 1. However, in most inland regions of northern Australia the dry seasons are usually longer, and there is a lower probability of any significant winter rainfall events than at the experimental site. This suggests that the results observed in the present study during the years with the lower rainfall but typical rainfall distribution (Expt 1a and Expt 2) are likely to be most representative of the northern Australian cattle industry.
The large variation between years in the present study was especially apparent with respect to cow LW change. Responses in PIP to improved P intake (Figure 1b) were apparent in 4 of the 5 grazing cycles but were most marked in the 2 years of lower rainfall (Expts 1a and 2). The between-year differences in PIP profiles of the Pdefic treatment indicated that the severity of dietary P deficiency was not constant across grazing cycles, although PIP concentrations did decline to <20 mg P/L during some intervals in 4 of the 5 annual cycles (the exception being Expt 1b, 1988-89). Thus the breeders in the Pdefic treatment were often severely deficient in dietary P (Wadsworth et al. 1990; Coates 1994; Anderson et al. 2017), while the PIP concentrations of 50‒60 mg P/L in the Psupp treatment established that these breeders were ingesting adequate dietary P for their level of production.
The poor LW performance of the Pdefic cows compared with the Psupp cows was most marked in Expts 1a and 2, with mean differences between treatments in annual LW change of 92 and 115 kg, respectively. These large responses to improved dietary P were comparable with large LW responses to P supplement of 62 and 88 kg in breeder cows also grazing grass-stylo pastures during 2 annual cycles in another experiment at a P-deficient site at Springmount, Mareeba, in the seasonally dry tropics of northern Australia (Miller et al. 1998; Dixon et al. 2016; Coates et al. 2018). In Expt 2 additional dietary P was provided via applied P fertilizer, which potentially had additional dietary advantages besides improved P nutrition. However, previous work with similar stylo-grass pastures and low-P soil at the same location showed that annual LW gain was the same for yearling heifers grazing either pasture fertilized annually with superphosphate or grazing unfertilized pasture but supplemented with P (Coates 1994). It therefore appears unlikely that in the present study factors other than dietary P contributed in any substantial way to differences in productivity between the Psupp and Pdefic treatments in Expt 2.
We hypothesize that the variation in responses to additional dietary P between annual cycles was due primarily to variation in the amounts and distribution of rainfall (i.e. seasonal conditions) with consequences for P concentration in the diets selected. The absence of treatment effects on cow LW changes during Expt 1b was most likely due to the favorable seasonal conditions throughout Expt 1b with well-above-average rainfall and an extended period of pasture growth through March and April, i.e. extended green season as defined by McCown (1981). This was in marked contrast to the responses during Expt 1a and Expt 2. Since P concentration is highest in young green leaf and lowest in mature, dry forage, an extended green season will promote higher P intakes through the grazing cycle (Coates 1994). Therefore, favorable seasonal conditions during Expt 1b probably resulted in better-than-average dietary P intakes, especially during lactation. This was supported by the higher PIP concentrations of the Pdefic cows during Expt 1b, i.e. mean PIP concentration during the December‒June lactation interval (41 mg P/L) was appreciably higher than the mean concentrations of 22, 26 and 14 mg P/L for Expts 1a, 1c and Expt 2, respectively (Figure 1b). In Expt 1d the PIP concentrations for the Pdefic treatment cows remained high (>40 mg P/L, mean 52 mg P/L from July to January) for much of the grazing cycle and declined to much lower levels only late in the grazing cycle. This may have been a consequence of the high rainfall from December 1990 to February 1991 in this experiment (Table 1).
The observation in the present study that the adverse effects of P deficiency in breeder cows were much greater in low-rainfall years is in accord with past observations in the northern Australian rangelands with low-P soils. The symptoms of acute P deficiency such as ‘peg-leg’ and reduced breeder productivity have been observed to occur most often during droughts and low-rainfall years (Turner et al. 1935; Rose 1954; Barnes and Jephcott 1955, 1959; Ferguson and Sklan 2005). Similarly, a 5-year experiment in semi-arid New Mexico, USA, examining effects of P supplementation on breeder cow performance, reported that the absence of P supplementation had a detrimental effect only when coupled with the effects of drought (Judkins et al. 1985). In 2 studies reported by Holroyd et al. (1977; 1983) with breeders grazing native grass pasture or native grass-stylo pastures, there was relatively low variation among years in the LW responses of breeders to P supplements fed during the wet and early dry seasons at a site in the seasonally dry tropics of northern Australia similar to that at Lansdown. However this was associated with low variation in both total rainfall and the rainfall pattern. We suggest, therefore, that the present study is in accord with these other reported studies in demonstrating a large impact of variable seasonal conditions on the LWs and fertility of breeder cows grazing pastures on low-P soils.
The magnitude of the responses in Pdefic cows to the provision of additional dietary P in the present study was directly related to the extent of the annual LW losses in the Pdefic cows (Figure 4). The largest LW responses to additional dietary P occurred during those annual cycles when the Pdefic cows lost the greatest LW. Furthermore, large LW responses of Pdefic cows to additional dietary P in 2 drafts of another comparable study with similar genotype breeders in northern Queensland (Miller et al. 1998; Dixon et al. 2016; Coates et al. 2018) were consistent with this relationship. Presumably the annual LW gain of Psupp cows in the present study was constrained by the availability of other nutrients such as protein and energy rather than a lack of P through the annual cycle. The magnitude of the cow LW responses to P supplement was also related to the PIP of the Pdefic cows during lactation; these PIP concentrations were, except for Expt 1b, in the range 14‒26 mg P/L indicating severe P deficiency (Wadsworth et al. 1990; Coates 1994; Anderson et al. 2017). The high PIP concentrations (50‒59 mg P/L) in the Psupp cows in all annual cycles established that the breeders in this treatment were ingesting adequate dietary P. Thus the variations in the responses to P supplement were closely linked to the amount and pattern of rainfall during each annual cycle and to the performance of the Pdefic breeders. However, the magnitude of the variations in the responses of Pdefic breeders to P supplements in the present study and in that of Miller et al. (1998), where the breeders also grazed stylo-grass pastures growing on P-deficient soils, may have been greater than would occur with breeders grazing native grass pastures containing little or no legume. Stylo-grass pastures growing on low-P soils are usually higher in protein and metabolizable energy relative to P concentration than grass pastures, and also have a high Ca:P ratio (e.g. often >10:1 and up to 25:1), which reduces bone-P mobilization (Minson 1990; Coates 1994). Consequently, the responses of growing cattle to supplementary P will often be much higher for stylo-grass pastures than for grass pastures. This may also occur with breeders grazing grass-stylo pastures. A lesser variation between annual cycles in the LW response of breeders grazing grass pastures to P supplements would presumably lead to a lesser variation in other measures of breeder herd productivity.
Figure 4 The relationship in paddock groups of cows between the increase in cow LW due to the provision of P supplements and the LW change of the Pdefic cows during the annual cycle for groups of cows in the present study (o) and for 2 annual drafts of an experiment (●) at Mareeba, north Queensland, Australia, where breeder cows of similar genotype grazed P-deficient pasture and received no supplement or were supplemented with P (Miller et al. 1998). The regression of the results from the present study was: Y = 19.9 − 1.29 X (R2 = 0.93; P<0.01). The regression of the pooled results was: Y = 20.3 − 1.06 X (R2 = 0.84; P<0.01).
An important observation in the present study was that calf growth was generally similar in both unsupplemented and supplemented groups of breeders. Thus the breeders must have largely maintained milk production, regardless of the dietary P deficiency, even when undergoing substantial LW loss through annual cycles. The LW gain of calves suckling Pdefic cows was on average only 0.09 kg/day (or c. 10%) lower than that of calves suckling Psupp cows, a difference of generally lower economic importance than cow fertility and mortality in rangeland production systems (M. Bowen and F. Chudleigh pers. comm.). The Bos indicus × Bos taurus genotype cows used in the present study must have been able to mobilize body reserves of both energy and P to maintain lactational output when nutritional intake was insufficient for LW maintenance. However, such mobilization of body reserves by lactating cows must clearly depend on availability of sufficient body reserves for this purpose. The Pdefic cows in the present study apparently had sufficient body reserves for this purpose but obviously P mobilization can continue only until body P reserves become depleted. The adverse effects of exhaustion of body P reserves with large and severe effects in the second and third years of ongoing P deficiency on intake, LW and milk production have been shown by Read et al. (1986) for Bonsmara crossbred beef cows and by Valk and Sebek (1999) for dairy cows. If the cows used in Expt 1a or Expt 2 had become pregnant and continued for another annual cycle of severe P deficiency, then poor lactational performance and/or cow mortality would probably have occurred. When cows mobilize body P and body energy reserves (as LW and body condition) during early lactation it is obviously essential that the reserves be replaced in late lactation or post-weaning for annual calving in harsh and P-deficient seasonally dry tropical environments (Dixon et al. 2017).
Although the number of cows in each treatment during each annual cycle was small in the present study, the pronounced effects of Pdefic on pregnancy rates in cows in Expts 1a and 2, but the absence of differences in Expts 1b and 1c (not measured in Expt 1d), indicate the differences likely to occur in commercial herds. The very poor reconception rates in the Pdefic cows in Expts 1a and 2 were associated with low cow LW and body condition throughout the mating period. Breeder cows grazing P-deficient pastures are often prone to lactational anoestrus and fail to conceive while lactating, resulting in patterns of either delays in re-conception in consecutive years (i.e. ‘slippage’) or calving every second year (e.g. Read et al. 1986; McGowan et al. 2014; Dixon et al. 2017). Such a calving pattern was not observed in the Pdefic cows during the grazing cycles of Expts 1b, 1c and 1d but was probably a consequence of the sequence of annual cycles with well-above-average rainfall (Table 1).
Conclusions
In breeders grazing P-deficient Stylosanthes-grass pastures at a site in a seasonally dry environment, there was wide variation among annual cycles in the severity of P deficiency and the LW performance and LW responses of reproducing breeders to provision of adequate dietary P. This was apparently associated with variation in seasonal conditions and dietary P intakes of un- supplemented animals. Thus the assessment of the severity and economic consequences of P deficiency on breeder performance within a given paddock requires cognizance of potential between-year differences and may require diagnostic measurements over a number of annual grazing cycles that encompass a range of seasonal conditions.
